Understanding molecular variations of cattle genome by machine learning

With large-scale omics data available, this PhD project will train a PhD researcher with innovative ideas of integrating informatics into animal genetic and genomic research in the new era. The student will integrate genetics and machine learning methods to identify molecular variations of the cattle genome and assess the association of molecular variants with complex phenotypes (e.g., health, production, efficiency, fitness).

The successful candidate will have the opportunity to work with large-scale cattle omics data including whole-genome sequence, RNA-sequence, complex phenotypes of UK cattle population. In addition to the data, the student will have access to methodology, software, and high-performance computational resources (e.g., Nvidia DGX Workstation, clusters, Jupyter Notebook) that have been well-established by us to address the research questions.

The data will be jointly analyzed to identify molecular variations of cattle genome (e.g., eQTL) using sequence data and machine/deep learning methods. The identified variants with functional significance will be associated with cattle complex phenotypes to investigate the links between functional variants and cattle health, fitness, production, and efficiency.

The PhD project offers interdisciplinary research and the project has strong collaborative efforts between SRUC and University of Edinburgh through cross-supervision.

The student will obtain fruitful research training from both SRUC and University of Edinburgh on quantitative genetics, statistical modelling, and machine learning. The student will also obtain valuable training in scientific writing, project management skills, and career development skills. Students are part of the EASTBIO training programme and will undertake enhanced subject-specific, core bioscience and generic skills training and a 3-month professional internship (PIPS) outwith academia. The students are expected to submit a PhD thesis in 4 years.